Science fiction, like *Star Wars* has long portrayed robots such as C-3PO and R2-D2 with human-level or superior intelligence. According to Tony Prescott, a professor of Cognitive Robotics, this scenario might not be far from reality. In his book, *The Psychology of Artificial Intelligence* (Routledge), Prescott argues that combining multiple AIs into a larger, interconnected system—similar to the structure of the human brain—could be the key to AI surpassing human intelligence.
Prescott explains that various AIs, each specialized in different tasks, could be linked together to form a unified intelligence. When embodied in a robot, this system could enable AI to understand its surroundings, learn like humans, and become self-aware. This would allow AI to act intelligently in response to its environment.
While some argue that AI might never match human intelligence, recent advancements in AI, such as deep learning and generative models like ChatGPT, have accelerated the conversation. However, these models still lack true understanding—they excel at tasks like text generation but don’t comprehend the meaning behind their outputs.
For AI to match or exceed human intelligence, Prescott suggests that it would need to interact with the world through a robotic body, gaining real-world experience and awareness. The next step towards human-level intelligence might involve connecting different types of AI, like vision-based AIs, to create a cognitive architecture that mirrors the human brain.
Despite breakthroughs, aspects of human intelligence—such as creativity and imagination—remain out of reach for AI. According to Prescott, only by giving AI a body and the ability to physically experience its environment can it truly understand the world and potentially match human cognitive abilities.
